The liability method would trump FASB's compromise

Our two preceding columns exposed the weak arguments against expensing options, and then analyzed the incomplete methods that predated the Financial Accounting Standards Board’s March 2004 exposure draft. This one compares the proposed method with what we call the “liability method.”

The primary flaw in FASB’s proposal is its reliance on the old-fashioned matching model that was rejected in the board’s Conceptual Framework, because it calculates expense without paying attention to what happens to the real values of real assets and liabilities. As a result, FASB’s method will misrepresent the options’ total cost (equal to the difference between the strike price and the stock’s market value on the exercise date), fail to clearly unveil the dilutive effects of the option overhang, report more income tax expense than the company actually pays, and understate the equity created when the stock is issued.

Although FASB’s proposal takes the huge political step of mandating option expense, it barely tweaks SFAS 123’s compromised method, and thus falls far short of reporting useful information to the capital markets. Of course, when those markets are fully informed, everyone benefits because capital is priced to reflect actual risk/return tradeoffs.

In contrast to FASB’s method, the foundation of our proposed solution is the premise that stock options are derivative liabilities. Here is the crux of our thinking: Issuing options creates a present obligation to sell shares in the future at a price lower than their then-prevailing market value. As we interpret the Conceptual Framework, present obligations to make future sacrifices are liabilities, whether the sacrifice involves paying cash, surrendering a non-cash asset, providing services or issuing a company’s own shares at a discount.

Our premise contradicts conventional assumptions that option grants create ownership interests and thus cause subsequent changes in their value to be unrecognized, just like value changes for outstanding shares.

We fault this traditional view for failing to acknowledge this crucial point: Option holders do not have the full rights, risks and rewards of owners. Specifically, option holders do not receive dividends, cannot vote, cannot receive assets upon liquidation, and have risks that are different from those faced by shareholders. Thus, we have come to the vastly different, but quite sound, conclusion that options are liabilities.

This view leads to more useful information by reporting changes in option value after the grant date. These changes can be significant, because options are immensely risky and volatile. Issuers face the risk that the value could climb sky-high and leave them collecting but a fraction of the optioned shares’ market value. Employees face the risk that the options could go underwater and stay there, leaving them with nothing. According to the framework, information is reliable only when it is representationally faithful; thus, any method of accounting for options will produce reliable results only if it fully reflects those risks and the resulting volatility.

On that test alone, FASB’s proposed method would publish unreliable information, because it won’t require management to report options (as liabilities or equity) on the grant date balance sheet at any amount. Although the grant date value eventually appears in equity after the vesting period, that amount is no longer timely nor otherwise representative of current conditions. Consequently, the board’s method provides neither relevant nor reliable information, thus perpetuating the need for the capital markets and managers to keep looking elsewhere for useful facts.

What would be the results of accounting for options as derivative liabilities? Pretty useful, we think.
For one thing, continuously marking the options liability to market ensures that its full amount appears on the balance sheet, thereby describing the economic effects of issuing a risky obligation without a fixed maturity value. When the options’ value goes up, the liability balance is credited and the offsetting debit is an expense (or a loss) arising from compensating employees with options.

On the other hand, when the value decreases, the liability is debited and the offsetting credit is an expense reduction (or a gain). Note that these volatile changes in real value are reported on the income statement as soon as they occur. The recognized numbers are not dead-and-gone past facts, but current verifiable observations of today’s facts. This timely and faithfully representative information is simply more useful.

The following illustration shows the differences between these methods. Suppose a company issues nonqualified 10-year options that vest in two years and have zero intrinsic value at the grant date but a $5 million market value. The holders can acquire one million shares for $10 each. Ten years later, the options are exercised at a $50 million discount below the stock’s market value. Applying the company’s 40 percent tax rate produces a tax savings of $20 million at the exercise date.

Following the grandfathered method of APBO 25, management never reports compensation expense; never acknowledges the options on the balance sheet; reports income tax expense over the 10 years that exceeds taxes paid by $20 million; and credits the common stock account for $10 million, well short of the stock’s $60 million value.

FASB’s latest proposal, while labored over for a decade, does only a little better. Over the options’ life, the issuer recognizes total expense of only $5 million (spread over the initial two-year vesting period); reports only a $5 million equity item on the balance sheet at the end of year two and beyond, even though the options’ value rises much higher; reports tax expense over the 10 years that exceeds taxes paid by $18 million (40 percent of the $45 million unreported compensation expense); and credits the common stock account for $15 million at exercise, well short of the stock’s $60 million value.

But look at what happens when options are treated as a liability. Over the options’ life, the issuer reports the full $50 million cost, recognized partially in each year in accordance with observed changes in option value instead of predictions conjured on the grant date; publishes up-to-date balance sheets with a liability equal to the options’ current value (not the long-past grant date value); recognizes total tax expense over the 10-year life that equals the total taxes paid; and credits the common stock account for the full $60 million value of the issued shares. The raw and unsmoothed volatility in reported income over the options’ life reflects actual volatility, which makes the income statement reliable and useful.

What if the options go underwater? Let’s apply the same facts except that the options lapse at the end of 10 years with a value of $0.

The pathetic APBO 25 method causes accountants, like horses wearing blinders, to prepare financial statements as if the options don’t exist, even though a cascade of events is happening all around them, including the acquisition of labor, the issuance of derivative instruments, and the decline in the options’ value. All is blissfully quiet and uneventful, or so it appears.

As we analyze FASB’s proposal, we find it to be inadequate because it also disregards most of the relevant events. Management reports the same $5 million compensation in the first two years but continues to carry the options account at $5 million up to the lapsing date, never revealing (until the end) that the options’ value is changing and, eventually, disappearing. Management at least gets to report income tax expense that equals taxes paid and a zero balance for common stock that wasn’t issued.

Bottom line, except for the $5 million expense allocated to the first two years, FASB’s new method produces virtually the same results as APBO 25. More condemning is the fact that it produces essentially the same result as when the options went through the roof. It’s clear to us that FASB’s best effort does not produce comparable or useful information because it doesn’t distinguish between even these grossly different scenarios.

The liability method would cause management to report annual expense effects describing the change in the options’ value (eventually netting to zero); recognize a liability that would vary but eventually converge on zero; report income tax expense over 10 years equal to the taxes paid; and make no entry to the common stock account. It tells the whole story, even if no one expected it to turn out this way.

Although we need more space to explain more completely, we think we have illustrated the advantages of treating options as derivative liabilities. By “advantages,” of course, we mean that the capital markets (and managers) are provided with more useful information.

We also want to make it clear that we really don’t expect FASB to discard its proposal and adopt the liability method instead. Too many strident battles have drained the board’s political energy and resources. It is time, perhaps, for the board to declare victory and then retreat as fast as possible.

Because we are still optimistic, we have privately proposed to FASB that it can enable more informed assessments by recommending that managers disclose the options’ current market value on each balance sheet date. Those who adopt this practice would declare their willingness to serve financial statement users’ needs. To these voluntarily transparent few would go the spoils of lower capital costs and higher stock prices.

We suspect, however, that most managers will not have a clue. After all, that condition has been the ultimate source of all the stock option problems since the beginning.

For reprint and licensing requests for this article, click here.
MORE FROM ACCOUNTING TODAY